Skip to main content

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

  • Article
  • Published:

Anchoring and synaptic stability of PSD-95 is driven by ephrin-B3

Nature Neuroscience volume 18, pages 1594–1605 (2015)Cite this article

Subjects

Abstract

Organization of signaling complexes at excitatory synapses by membrane-associated guanylate kinase (MAGUK) proteins regulates synapse development, plasticity, senescence and disease. Post-translational modification of MAGUK family proteins can drive their membrane localization, yet it is unclear how these intracellular proteins are targeted to sites of synaptic contact. Here we show using super-resolution imaging, biochemical approaches and in vivo models that the trans-synaptic organizing protein ephrin-B3 controls the synaptic localization and stability of PSD-95 and links these events to changes in neuronal activity via negative regulation of a newly identified mitogen-associated protein kinase (MAPK)-dependent phosphorylation site on ephrin-B3, Ser332. Unphosphorylated ephrin-B3 was enriched at synapses, and interacted directly with and stabilized PSD-95 at synapses. Activity-induced phosphorylation of Ser332 dispersed ephrin-B3 from synapses, prevented the interaction with PSD-95 and enhanced the turnover of PSD-95. Thus, ephrin-B3 specifies the synaptic localization of PSD-95 and likely links the synaptic stability of PSD-95 to changes in neuronal activity.

This is a preview of subscription content, access via your institution

Access options

Buy this article

  • Purchase on Springer Link
  • Instant access to full article PDF
Buy now

Prices may be subject to local taxes which are calculated during checkout

Figure 1: PSD-95 is in a complex with ephrin-B3 at synapses.
Figure 2: Ephrin-B3 links MAPK signaling to PSD-95–ephrin-B3 interaction and localization.
Figure 3: ERK phosphorylation of Ser332 regulates subcellular localization of ephrin-B3.
Figure 4: Ephrin-B3 regulates subcellular PSD-95 localization.
Figure 5: Organization and localization of ephrin-B3 is independent of PSD-95.
Figure 6: Ephrin-B3 regulates PSD-95 mobility.
Figure 7: Ephrin-B3 regulation of PSD-95 mobility requires ERK-binding D-domain and Ser332.
Figure 8: MAPK negatively regulates PSD-95 mobility.

Similar content being viewed by others

References

  1. Feng, W. & Zhang, M. Organization and dynamics of PDZ-domain-related supramodules in the postsynaptic density. Nat. Rev. Neurosci. 10, 87–99 (2009).

    Article  CAS  PubMed  Google Scholar 

  2. Huganir, R.L. & Nicoll, R.A. AMPARs and synaptic plasticity: the last 25 years. Neuron 80, 704–717 (2013).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  3. Ehrlich, I., Klein, M., Rumpel, S. & Malinow, R. PSD-95 is required for activity-driven synapse stabilization. Proc. Natl. Acad. Sci. USA 104, 4176–4181 (2007).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  4. Elias, G.M. & Nicoll, R. Synaptic trafficking of glutamate receptors by MAGUK scaffolding proteins. Trends Cell Biol. 17, 343–352 (2007).

    Article  CAS  PubMed  Google Scholar 

  5. El-Husseini, A., Schnell, E., Chetkovich, D., Nicoll, R. & Bredt, D. PSD-95 involvement in maturation of excitatory synapses. Science 290, 1364–1368 (2000).

    CAS  PubMed  Google Scholar 

  6. Bayés, A. et al. Characterization of the proteome, diseases and evolution of the human postsynaptic density. Nat. Neurosci. 14, 19–21 (2011).

    Article  CAS  PubMed  Google Scholar 

  7. Grant, S.G. Synaptopathies: diseases of the synaptome. Curr. Opin. Neurobiol. 22, 522–529 (2012).

    Article  CAS  PubMed  Google Scholar 

  8. Blanpied, T.A., Kerr, J. & Ehlers, M. Structural plasticity with preserved topology in the postsynaptic protein network. Proc. Natl. Acad. Sci. USA 105, 12587–12592 (2008).

    Article  PubMed  PubMed Central  Google Scholar 

  9. Gray, N.W., Weimer, R., Bureau, I. & Svoboda, K. Rapid redistribution of synaptic PSD-95 in the neocortex in vivo. PLoS Biol. 4, (2006).

  10. Kuriu, T., Inoue, A., Bito, H., Sobue, K. & Okabe, S. Differential control of postsynaptic density scaffolds via actin-dependent and -independent mechanisms. J. Neurosci. 26, 7693–7706 (2006).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  11. Sturgill, J., Steiner, P., Czervionke, B. & Sabatini, B. Distinct domains within PSD-95 mediate synaptic incorporation, stabilization, and activity-dependent trafficking. J. Neurosci. 29, 12845–12854 (2009).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  12. Ho, G.P. et al. S-nitrosylation and S-palmitoylation reciprocally regulate synaptic targeting of PSD-95. Neuron 71, 131–141 (2011).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  13. Nelson, C., Kim, M., Hsin, H., Chen, Y. & Sheng, M. Phosphorylation of threonine-19 of PSD-95 by GSK-3β is required for PSD-95 mobilization and long-term depression. J. Neurosci. 33, 12122–12135 (2013).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  14. Steiner, P. et al. Destabilization of the postsynaptic density by PSD-95 serine 73 phosphorylation inhibits spine growth and synaptic plasticity. Neuron 60, 788–802 (2008).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  15. Kim, M.J. et al. Synaptic accumulation of PSD-95 and synaptic function regulated by phosphorylation of serine-295 of PSD-95. Neuron 56, 488–502 (2007).

    Article  CAS  PubMed  Google Scholar 

  16. Craven, S.E., El-Husseini, A. & Bredt, D. Synaptic targeting of the postsynaptic density protein PSD-95 mediated by lipid and protein motifs. Neuron 22, 497–509 (1999).

    Article  CAS  PubMed  Google Scholar 

  17. Biederer, T. & Stagi, M. Signaling by synaptogenic molecules. Curr. Opin. Neurobiol. 18, 261–269 (2008).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  18. Zheng, C.-Y., Seabold, G., Horak, M. & Petralia, R. MAGUKs, synaptic development, and synaptic plasticity. Neuroscientist 17, 493–512 (2011).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  19. Mah, W. et al. Selected SALM (synaptic adhesion-like molecule) family proteins regulate synapse formation. J. Neurosci. 30, 5559–5568 (2010).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  20. Shipman, S.L. et al. Functional dependence of neuroligin on a new non-PDZ intracellular domain. Nat. Neurosci. 14, 718–726 (2011).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  21. Hruska, M. & Dalva, M. Ephrin regulation of synapse formation, function and plasticity. Mol. Cell. Neurosci. 50, 35–44 (2012).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  22. McClelland, A.C., Hruska, M., Coenen, A., Henkemeyer, M. & Dalva, M. Trans-synaptic EphB2-ephrin-B3 interaction regulates excitatory synapse density by inhibition of postsynaptic MAPK signaling. Proc. Natl. Acad. Sci. USA 107, 8830–8835 (2010).

    Article  PubMed  PubMed Central  Google Scholar 

  23. Kayser, M., McClelland, A., Hughes, E. & Dalva, M. Intracellular and trans-synaptic regulation of glutamatergic synaptogenesis by EphB receptors. J. Neurosci. 26, 12152–12164 (2006).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  24. Chen, X. et al. PSD-95 is required to sustain the molecular organization of the postsynaptic density. J. Neurosci. 31, 6329–6338 (2011).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  25. Hafner, A.-S.S. et al. Lengthening of the stargazin cytoplasmic tail increases synaptic transmission by promoting interaction to deeper domains of PSD-95. Neuron 86, 475–489 (2015).

    Article  CAS  PubMed  Google Scholar 

  26. Wollscheid, B. et al. Mass-spectrometric identification and relative quantification of N-linked cell surface glycoproteins. Nat. Biotechnol. 27, 378–386 (2009).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  27. Kayser, M.S., Nolt, M. & Dalva, M. EphB receptors couple dendritic filopodia motility to synapse formation. Neuron 59, 56–69 (2008).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  28. Thomas, G.M. & Huganir, R. MAPK cascade signalling and synaptic plasticity. Nat. Rev. Neurosci. 5, 173–183 (2004).

    Article  CAS  PubMed  Google Scholar 

  29. Lau, L.F. et al. Interaction of the N-methyl-D-aspartate receptor complex with a novel synapse-associated protein, SAP102. J. Biol. Chem. 271, 21622–21628 (1996).

    Article  CAS  PubMed  Google Scholar 

  30. Holtmaat, A. & Svoboda, K. Experience-dependent structural synaptic plasticity in the mammalian brain. Nat. Rev. Neurosci. 10, 647–658 (2009).

    Article  CAS  PubMed  Google Scholar 

  31. Antion, M.D., Christie, L., Bond, A., Dalva, M. & Contractor, A. Ephrin-B3 regulates glutamate receptor signaling at hippocampal synapses. Mol. Cell. Neurosci. 45, 378–388 (2010).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  32. Rodenas-Ruano, A., Perez-Pinzon, M., Green, E., Henkemeyer, M. & Liebl, D. Distinct roles for ephrinB3 in the formation and function of hippocampal synapses. Dev. Biol. 292, 34–45 (2006).

    Article  CAS  PubMed  Google Scholar 

  33. Cane, M., Maco, B., Knott, G. & Holtmaat, A. The relationship between PSD-95 clustering and spine stability in vivo. J. Neurosci. 34, 2075–2086 (2014).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  34. Yoshii, A. & Constantine-Paton, M. BDNF induces transport of PSD-95 to dendrites through PI3K-AKT signaling after NMDA receptor activation. Nat. Neurosci. 10, 702–711 (2007).

    Article  CAS  PubMed  Google Scholar 

  35. Wilhelm, B.G., Groemer, T.W. & Rizzoli, S.O. The same synaptic vesicles drive active and spontaneous release. Nat. Neurosci. 13, 1454–1456 (2010).

    Article  CAS  PubMed  Google Scholar 

  36. Ehrlich, I. & Malinow, R. Postsynaptic density 95 controls AMPA receptor incorporation during long-term potentiation and experience-driven synaptic plasticity. J. Neurosci. 24, 916–927 (2004).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  37. Migaud, M. et al. Enhanced long-term potentiation and impaired learning in mice with mutant postsynaptic density-95 protein. Nature 396, 433–439 (1998).

    Article  CAS  PubMed  Google Scholar 

  38. Fernández, E. et al. Targeted tandem affinity purification of PSD-95 recovers core postsynaptic complexes and schizophrenia susceptibility proteins. Mol. Syst. Biol. 5, 269 (2009).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  39. Barrow, S.L. et al. Neuroligin1: a cell adhesion molecule that recruits PSD-95 and NMDA receptors by distinct mechanisms during synaptogenesis. Neural Dev. 4, 17 (2009).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  40. Giannone, G. et al. Neurexin-1β binding to neuroligin-1 triggers the preferential recruitment of PSD-95 versus gephyrin through tyrosine phosphorylation of neuroligin-1. Cell Reports 3, 1996–2007 (2013).

    Article  CAS  PubMed  Google Scholar 

  41. Engert, F. & Bonhoeffer, T. Dendritic spine changes associated with hippocampal long-term synaptic plasticity. Nature (1999).

  42. Bosch, M. et al. Structural and molecular remodeling of dendritic spine substructures during long-term potentiation. Neuron 82, 444–459 (2014).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  43. Meyer, D., Bonhoeffer, T. & Scheuss, V. Balance and stability of synaptic structures during synaptic plasticity. Neuron 82, 430–443 (2014).

    Article  CAS  PubMed  Google Scholar 

  44. Kwon, H.-B.B. et al. Neuroligin-1-dependent competition regulates cortical synaptogenesis and synapse number. Nat. Neurosci. 15, 1667–1674 (2012).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  45. Woods, G.F., Oh, W.C., Boudewyn, L.C., Mikula, S.K. & Zito, K. Loss of PSD-95 enrichment is not a prerequisite for spine retraction. The J. Neurosci. 31, 12129–12138 (2011).

    Article  CAS  PubMed  Google Scholar 

  46. Wu, G.Y., Deisseroth, K. & Tsien, R. Spaced stimuli stabilize MAPK pathway activation and its effects on dendritic morphology. Nat. Neurosci. 4, 151–158 (2001).

    Article  CAS  PubMed  Google Scholar 

  47. Sun, Q. & Turrigiano, G.G. PSD-95 and PSD-93 play critical but distinct roles in synaptic scaling up and down. J. Neurosci. 31, 6800–6808 (2011).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  48. Makino, H. & Malinow, R. Compartmentalized versus global synaptic plasticity on dendrites controlled by experience. Neuron 72, 1001–1011 (2011).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  49. Frank, C.A., Pielage, J. & Davis, G.W. A presynaptic homeostatic signaling system composed of the Eph receptor, ephexin, Cdc42, and CaV2.1 calcium channels. Neuron 61, 556–569 (2009).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  50. Ebert, D.H. & Greenberg, M.E. Activity-dependent neuronal signalling and autism spectrum disorder. Nature 493, 327–337 (2013).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  51. McClelland, A.C., Sheffler-Collins, S., Kayser, M. & Dalva, M. Ephrin-B1 and ephrin-B2 mediate EphB-dependent presynaptic development via syntenin-1. Proc. Natl. Acad. Sci. USA 106, 20487–20492 (2009).

    Article  PubMed  PubMed Central  Google Scholar 

  52. Yokoyama, N. et al. Forward signaling mediated by ephrin-B3 prevents contralateral corticospinal axons from recrossing the spinal cord midline. Neuron 29, 85–97 (2001).

    Article  CAS  PubMed  Google Scholar 

  53. Leslie, K., Nelson, S. & Turrigiano, G. Postsynaptic depolarization scales quantal amplitude in cortical pyramidal neurons. J. Neurosci. 21, RC170 (2001).

  54. Lois, C., Hong, E., Pease, S., Brown, E. & Baltimore, D. Germline transmission and tissue-specific expression of transgenes delivered by lentiviral vectors. Science 295, 868–872 (2002).

    Article  CAS  PubMed  Google Scholar 

  55. Takasu, M.A., Dalva, M., Zigmond, R. & Greenberg, M. Modulation of NMDA receptor-dependent calcium influx and gene expression through EphB receptors. Science 295, 491–495 (2002).

    Article  CAS  PubMed  Google Scholar 

Download references

Acknowledgements

We would like to thank W. Zhou for comments on FRAP data analysis and the students and faculty of the Neurobiology course (2009–2010) at the Marine Biological Laboratory for help with pilot experiments. This work was supported by grants from NIDA (DA022727) and NIMH (MH086425 and MH100093) to M.B.D.

Author information

Author notes

  1. Sylvain J Le Marchand

    Present address: Present address: Cell Imaging Center, Drexel University, Philadelphia, Pennsylvania, USA.,

Authors and Affiliations

  1. Department of Neuroscience and the Farber Institute for Neuroscience, Thomas Jefferson University, Jefferson Hospital for Neuroscience, Philadelphia, Pennsylvania, USA

    Martin Hruska, Nathan T Henderson, Nan L Xia, Sylvain J Le Marchand & Matthew B Dalva

Authors
  1. Martin Hruska
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Nathan T Henderson
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Nan L Xia
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Sylvain J Le Marchand
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Matthew B Dalva
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

M.H. and M.B.D. designed the project. Biochemistry, live imaging and sensory deprivation were performed by M.H. Immunocytochemistry and imaging was performed by M.H., N.T.H., N.L.X. and S.J.L.M. Organotypic slice culture was performed by M.H. and N.T.H. Immunohistochemistry was performed by N.T.H. M.H., N.T.H., N.L.X. and S.J.L.M. analyzed the data. The manuscript was written by M.H. and M.B.D.

Corresponding author

Correspondence to Matthew B Dalva.

Ethics declarations

Competing interests

The authors declare no competing financial interests.

Integrated supplementary information

Supplementary Figure 1 Data related to figure 1: Selective interaction between ephrin-B3 and PSD-95 in synaptosomes and in vitro.

(a) Source of western blot data shown in Figure 1d (a, top right blot), 1e (a, top left blot), and 1f (a, bottom two blots). Blots show that ephrin-B3 and PSD-95 fail to co-IP from wild type (WT) and ephrin-B3 null (Efnb3–/–) whole brain lysates. Boxes indicate approximate regions show in the main figure. Probed as shown at the bottom of each row of blots.

(b) Source of western blot data shown in Figure 1g (b, top right blot), 1h (b, top left blot), and 1i (b, bottom two blots). Blots show ephrin-B3 and PSD-95 co-IP from P21 wild type (WT) but not ephrin-B3 null (Efnb3–/–) synaptosomes. Boxes indicate approximate regions show in the main figure.

(c) Source of western blot data shown in Figure 1j (c, top left blot), 1k (c, four top right blots), and 1l (c, bottom blots). Blots demonstrate specific co-IP of PSD-95 with ephrin-B3, but not ephrin-B1 or ephrin-B2 from P21 synaptosomes. Boxes indicate approximate regions show in the main figure.

(d) PSD-95-GST and full-length HA-tagged ephrin-B1-B3 were used for the pull-down experiments. Considerably higher pull-down signal was detected with HA-tagged ephrin-B3 than with HA-ephrin-B1 or HA-ephrin-B2. Middle blot shows the source of western loading control blot of HA-tagged ephrin-B proteins used in the binding assay.

(e) Source of western blot data shown in the pull-down (top blot panel S1d). Boxes indicate approximate regions show in the figure.

(f) Source of western blot data shown the flow-through (bottom blot panel S1d). Boxes indicate approximate regions show in the main figure.

Supplementary Figure 2 Data related to figure 2: Ephrin-B3-PSD-95 interaction is regulated by two distinct MAPK domains in the ephrin-B3 intracellular region.

(a) Source of the western blot data shown in IP experiments in figure 2b. Blots show co-IP experiments in HEK293T cells transfected with PSD-95-GFP and the indicated flag-tagged ephrin-B3 mutants. Boxes indicate approximate regions show in the main figure. Blots shown from top to bottom in figure 2b are shown in order from left to right.

(b) Source of the western blot data shown in lysate blots in figure 2b. Blots of inputs for the co-IP experiments in HEK293T cells transfected with PSD-95 GFP and the indicated flag-tagged ephrin-B3 constructs as shown in figure 2b. Boxes indicate approximate regions show in the main figure. Bottom two blots in figure 2b are shown in order from left to right.

(c) Source of the western blot data shown in figure 2d. Blots show pull-down experiments of in vitro translated ephrin-B3 intracellular domains (wild type and mutant) and the recombinant PSD-95-GST protein from figure 2d. Boxes indicate approximate regions show in the main figure.Blots shown from top to bottom in figure 2d are shown in order from left to right.

(d-e) Panels show results of molecular replacement experiments using ephrin-B3 PSD-95 binding muntants on synapse density in cortical neurons. Cortical neurons were co-transfected with tdTomato and either control or ephrin-B3 shRNA constructs at DIV0. Expression of ephrin-B3 was rescued by co-transfecting shRNA resistant flag-tagged wild type or the indicated ephrin-B3 mutant constructs along with ephrin-B3 shRNA. Neurons were fixed at DIV14 and stained with the indicated antibodies. Synapse density was measured as co-localization of PSD-95 and vGlut1 (arrows). Only neurons with at least 50 µm of tdTomato positive dendrites were included for the analysis. For the quantification of synapse density see figure 2t in the main text. Scale bar: 3 µm.

(f) Quantification of flag-tagged ephrin-B3 construct expression in neurons. Wild type and mutant (S332A and S332D) flag-tagged ephrin-B3 rescue constructs expressed at similar levels in neurons. The expression of the D-domain ephrin-B3 mutant (flag-eB3_L293A) was significantly higher, however, this mutant also exhibited significantly lower interaction with PSD-95 (Fig. 2 in the main text). Data are shown as mean ± s.e.m., p<0.0001, ANOVA, F (4, 129) = 6.903, with Tukey’s post hoc. ***p = 0.0004, **p = 0.0019, Control (n = 20), Wild type (n = 20), L293A (n = 32), S332D (n = 30), S332A (n = 32) neurons from at least three different transfections.

Supplementary Figure 3 Specificity of the phospho-Ser332 antibody.

(a) Alignment of cytoplasmic regions of all three ephrin-B family members, demonstrating a high degree of homology. Phosphorylable S332 residue found within a putative MAPK phosphorylation sequence PPQSPP is shown in the red box. Highlighted in green is the peptide that was used to generate and affinity purify the phospho-S332 (pS332) antibody.

The phospho-S332 (pS332) antibody appears to be selective because: 1) the antibody recognized ephrin-B3 in HEK 293T cells, but not ephrin-B1, ephrin-B2 or a non-phosphorylable ephrin-B3 (S332A) mutant (panel S3b); 2) the 37 kDa band corresponding to phospho-S332 is absent in brains from ephrin-B3 null mice (panel S3c); 3) treatment of neuronal lysates with calf intestinal phosphatase (CIP) resulted in the loss of the phospho-S332 band (panel S3d); and 4) immunostaining of cultured neurons following knockdown of ephrin-B3 (panels S3e-f) or of brain sections from Efnb3–/– mice (panels S3g-h) show significantly reduced levels of staining.

(b) Western blots showing that the phospho-S332 (pS332) antibody specifically recognizes WT ephrin-B3. Lysates from HEK 293T cells transfected with ephrin-B1 (HA-eB1), ephrin-B2 (HA-eB2), ephrin-B3 (Flag-eB3), or an ephrin-B3 with a mutation to S332 rendering it non-phosphorylatable (Flag-eB3_S332A). Only wild type ephrin-B3 is recognized by the pS332 antibody (arrow). Bottom blots were probed with α-HA and α-Flag to validate the expression of transfected ephrin-B constructs in HEK 293T cells.

(c) Western blots from WT and ephrin-B3 null (Efnb3–/–) cortical lysates probed with pS332 antibody. Arrow indicates the band of appropriate molecule weight is present only in WT lysates. The pS332 signal is absent from Efnb3–/– cortical lysates (arrow).

(d) Western blots of DIV14 cortical neuron lysates probed with the pS332 antibody treated with calf intestinal phosphatase (CIP, 1unit/µg of protein) for one hour. CIP treatment eliminates pS332 specific signal (arrow). Total ephrin-B3 expression (bottom blot, arrow) is not changed by CIP treatment.

(e) Representative images of dendritic segments from cortical neurons co-transfected with GFP and either control or ephrin-B3 shRNA constructs at DIV0. Neurons were stained with the pS332 antibody at DIV21. Arrows indicate pS332 puncta along GFP positive dendritic shaft. Sale bar: 3 µm.

(f) Knockdown of endogenous ephrin-B3 with shRNA causes a significant reduction in pS332 puncta density. Quantification of average density of pS332 ephrin-B3 puncta along dendrites of cultured neurons. Data are represented as the mean ± s.e.m., control (n = 5), ephrin-B3 shRNA (n = 6) neurons. ***p<0.001, un-paired t-test (two-tailed), t=9.181, df=9.

(g) P14 cortical sections from wild type and ephrin-B3 null (Efnb3–/–) mice stained with pS332 antibody. Prominent labeling is seen in wild type cortex, but this signal is absent in sections from ephrin-B3 null (Efnb3–/–) cortex. Scale bar: 200 µm.

(h) Representative images of apical dendrites of GFP labeled cortical neurons from wild type and ephrin-B3 null (Efnb3–/–) sections. Prominent pS332 labeling is observed in the shaft of wild type neurons, but this signal is absent in neurons from ephrin-B3 null brain. Scale bar: 3 µm.

Supplementary Figure 4 Activity-dependent regulation of Ser332 phosphorylation underlies ephrin-B3 localization.

(a) Source of the western blot data shown in figure 3b. Boxes indicate approximate regions show in the main figure. Probed as indicated at the bottom of each blot.

(b) Treatment of neurons with PD98059 blocks ephrin-B3 S332 phosphorylation. Western blots of neuronal lysates from DIV14 neurons treated with TTX (1 µM, 4 hours, Tocris) and MEK1/2 inhibitor PD98059 (100 µM, 4 hours, Sigma) as shown and probed with the indicated antibodies. 55 mM KCl (1 hour) was used to depolarize the neurons and induce Erk activation.

(c) Quantification of pS332 signal relative to total ephrin-B3. Depolarization induces a significant increase in pS332 signal that was blocked by pretreating neurons with PD98059. Data are shown as mean ± s.e.m., p = 0.0205, ANOVA, F (3, 8) = 5.843, with Fischer’s LSD post-hoc. *p = 0.0154 (TTX - KCl), **p = 0.0041 (TTX-PD - KCl), *p = 0.0298 (KCl - KCl+PD), n = 3 independent experiments.

(d) Source of the western blot data shown in panel S4b. Boxes indicate approximate regions show in the main figure. Probed as indicated at the bottom of each blot.

Supplementary Figure 5 Phosphorylated ephrin-B3 is localized outside synapses.

(a) Phospho-S332 is enriched in the shaft of cultured cortical neurons. DIV21 neurons transfected with GFP at DIV0 were fixed and stained with the indicated antibodies at DIV21. Arrows indicate pS332 puncta in the dendritic shaft and arrowheads point to dendritic spines. Scale bar: 3µm.

(b) Quantification of pS332 labeling in different cellular compartments per unit area of GFP labeled dendrite or dendritic spine. Phospho-S332 puncta were significantly less abundant in spines compared to dendritic shafts. Data are represented as mean ± s.e.m., ANOVA, F (2, 42) = 5.98, Tukey’s post-hoc. *p = 0.0467, **p = 0.005, n = 15 neurons from 2 independent experiments.

(c) Source of western blot data for the PSD fractionation experiments shown in figure 3e. Probed as indicated at the bottom of each blot. Blots shown top to bottom in figure 3e are shown in left to right order. Boxes indicate approximate regions show in the main figure.

(d) Unphosphorylated ephrin-B3 is enriched in mouse cortical synaptosomes. P21 wild type mouse cortex fractionated into indicated fractions and labeled with the indicated antibodies. Purification of synaptosomes was confirmed by the enrichment of PSD-95 in these fractions along with ephrin-B3.

(e) Quantification of the pS332 signal relative to the total ephrin-B3 expression indicates that a significantly higher proportion of phosphorylated ephrin-B3 is found outside of synapses. Ratios of pS332 to ephrin-B3 were normalized to S1 fractions. Data are represented as the mean ± s.e.m., p<0.0001, ANOVA, F (2, 6) = 62.87, with Tukey’s post-hoc. **p = 0.0002, ***p = 0.0001, n = 3 wild type mice.

(f) Source of western blots shown in panel S5d. Probed as indicated at the bottom of each blot. Boxes indicate approximate regions show in the main figure.

(g) Source of western blots for the synaptosomes prepared from control and activity deprived cortex shown in figure 3i. Probed as indicated at the bottom of each blot. Blots shown top to bottom in figure 3i are shown left to right. Boxes indicate approximate regions show in the main figure.

(h) Source of western blots for the ephrin-B3 and PSD-95 synaptosome co-IP experiments prepared from control and activity deprived cortical hemispheres as shown in figure 3j. Probed as indicated at the bottom of each blot. Blots shown top to bottom in figure 3j are shown left to right. Boxes indicate approximate regions show in the main figure.

Supplementary Figure 6 Synaptic localization of PSD-95 is reduced in ephrin-B3 null mice.

(a-c) The levels of synaptic CamKII (a), GluN2A (b), and GluN2B (c) are unchanged in mouse cortical synaptosomes from ephrin-B3 null (Efnb3–/–) animals. Synaptosomes were purified from cortices of P10 and P21 wild type and Efnb3–/– mouse brains and probed with antibodies to CamKII (a), GluN2A (b), and GluN2B (c) that are known to be synaptically localized. Probes for the western blots are shown at the bottom.

(d-f) Quantification of the synaptic enrichment of CamKII (d), GluN2A (e), and GluN2B (f) in Efnb3–/– mice relative to wild type (WT) at P10 and at P21. No difference in synaptic expression was detected in any of these synaptic proteins between wild type and ephrin-B3 null brains. Data are represented as mean ± s.e.m. Un-paired student’s t-test (two-tailed), p = 0.6185 (P10, CaMKII), p = 0.9382 (P21, CaMKII), p = 0.6750 (P10, GluN2A), p = 0.4971 (P21, GluN2A), p = 0.3223 (P10, GluN2B), p = 0.7510 (P21, GluN2B). N = 2 mice (WT, P10), n = 3 mice (Efnb3–/–, P10), n = 6 (WT, P21), n = 6 (Efnb3–/–, P21).

(g) Source of western blots for synaptosomes from P10 (left three blots) and P21 (right three blots) mouse brains, shown in figure 4a, demonstrating reduced PSD-95 enrichment in synaptosomes from ephrin-B3 null (Efnb3–/–) brains. Probed as indicated at the bottom of each blot.

Supplementary Figure 7 PSD-95 is less enriched in postsynaptic densities from Efnb3–/– brains.

(a-b) Biochemical fractionation of PSDs from the cortices of wild type (a) and ephrin-B3 null (Efnb3–/–, b) animals. Representative western blots probed with the indicated antibodies are shown. Ephrin-B3 is enriched in PSDs of wild type animals along with PSD-95 and NMDA glutamate receptor subunits. Enrichment of PSD-95 is significantly reduced in PSDs purified from Efnb3–/– brain. Blots were probed with antibodies shown on right.

(c) Quantification of PSD-95 enrichment in PSDs normalized to CaMKII expression levels. Data are represented as mean ± s.e.m., p = 0.0032 (t = 4.731 df = 6), un-paired t-test, two-tailed. N = 4 wild type mice, n = 4 Efnb3–/– mice.

(d) Source of western blots for PSD fractionation experiments shown in S7a and S7b. Top panels show data for S7a, bottom panels show data for S7b. Probed as indicated at the bottom of the second row of blots. Boxes indicate approximate regions show in the figure.

Supplementary Figure 8 Synaptic localization of PSD-95-GFP in organotypic brain slices and cultured neurons.

(a) Representative images of cortical neuron apical dendrites in organotypic brain slices. Images show examples of wild type, ephrin-B3 null (Efnb3–/–), Efnb3–/– + flag-ephrin-B3 rescue (flag-eB3) slices, or wild type slices transfected with shRNA targeting ephrin-B3, and shRNA plus flag-eB3 rescue. Neurons were biolistically transfected with the indicated constructs (Transfections) along with PSD-95-GFP and tdTomato. Organotypic slices were fixed 4 days after transfection, re-sectioned, and immunostained for the presynaptic marker vGlut1. Arrows indicate PSD-95-GFP puncta that co-localize with vGlut1 puncta; open arrowheads indicate PSD-95-GFP puncta that do no co-localize with vGlut1 puncta. Scale bar: 2 µm.

(b) Quantification of the fraction of PSD-95-GFP puncta that co-localize with vGlut1 puncta in the transfection conditions indicated in S8a. The fraction of PSD-95-GFP puncta that co-localize with vGlut1 puncta is not significantly different regardless of the presence or absence of ephrin-B3. Data are represented as mean ± s.e.m., p = 0.1522, one-way ANOVA, F (4, 73) = 1.732. Control, n = 14; Control (Efnb3–/–), n = 9; flag-eB3 (Efnb3–/–), n = 16; eB3 shRNA, n = 33; eB3 shRNA + flag-eB3, n = 6 neurons from at least three independent experiments.

(c) PSD-95-GFP appears punctate in DIV0 rat cortical neurons co-transfected with PSD-95-GFP and either control or EphB2 shRNA constructs. At DIV10 neurons were imaged and the density and organization of PSD-95-GFP puncta were analyzed.

(d) Quantification of the density of PSD-95-GFP puncta in neurons transfected with PSD-95-GFP and either control or EphB2 shRNA. EphB2 knockdown caused a significant reduction in the density of PSD-95-GFP puncta. ***p < 0.0001, t = 6.199 df = 37.

(e) Quantification of the intensity of GFP puncta in neurons transfected with PSD-95-GFP and either control or EphB2 shRNA. PSD-95-GFP formed distinct puncta in both control and EphB2 shRNA transfected neurons that did not differ in intensity, p = 0.2059, t = 1.288 df = 37.

(f) Quantification of the intensity of GFP in dendritic shafts of neurons transfected with PSD-95-GFP and either control or EphB2 shRNA. EphB2 shRNA did not cause a detectable increase in GFP fluorescence in dendritic shafts, p = 0.9903, t=0.01230 df = 37.

(g) Quantification of the ratio of the intensity of GFP in dendritic shafts to puncta in neurons transfected with PSD-95-GFP and control or EphB2 shRNA. Ratio of GFP fluorescence in the shaft relative to the GFP fluorescence in puncta was not significantly different between control and EphB2 transfected neurons, p = 0.3736, t = 0.9006 df = 37. For graphs in S8d-g, data are represented as mean ± s.e.m. Control, n = 18 neurons; EphB2 shRNA, n = 21 neurons from two independent biological replicates. Un-paired t-test (two-tailed) was used to determine statistical significance.

Supplementary Figure 9 PSD-95 expression is reduced with shRNA targeting PSD-95 mRNA.  

(a) HEK 293T cells were co-transfected with PSD-95-GFP and either control or PSD-95 shRNA (Open Biosystems) constructs in indicated amounts. Expression of PSD-95 shRNA caused a robust knockdown of PSD-95-GFP that was significantly different from control at all concentrations tested. Transfection of HEK 293T cells with a shRNA-resistant PSD-95-GFP (PSD-95-GFP-R) construct rescued the effects of PSD-95 shRNA, indicating the specificity of the shRNA in targeting PSD-95. Actin (bottom blot) was used as a measure of equal protein loading.

(b) Quantification of PSD-95 shRNA efficacy (n = 3 independent western blots except for 1.0 μg shRNA (n = 2 western blots)). Data are represented as mean ± s.e.m., p = 0.0011, F (4, 9) = 12.39, ANOVA with Dunnet’s post-hoc. **p = 0.0021 (0.25 µg), ***p = 0.0005 (0.5 µg), **p = 0.0025 (0.75 µg), **p = 0.0030 (1.0 µg).

(c) Validation of PSD-95 shRNA in cultured cortical neurons. Rat cortical neurons co-transfected with tdTomato and either control or PSD-95 shRNA vectors at DIV0. The specificity of shRNA on knocking down of endogenous PSD-95 was tested by co-transfecting neurons with a PSD-95-GFP-R construct along with PSD-95 shRNA. Neurons were fixed and stained with the indicated antibodies at DIV10. Representative images demonstrate a significant reduction of PSD-95 puncta in neurons transfected with PSD-95 shRNA compared to control. Expression of the PSD-95-GFP-R construct restored punctate PSD-95 expression along tdTomato labeled dendrites to control levels. Scale bar: 3µm.

(d) Quantification of PSD-95 knockdown in neurons from S9c. Data are represented as mean ± s.e.m., p = 0.0001, F (2, 28) = 12.81, ANOVA with Tukey’s post-hoc. **p = 0.0019, ***p = 0.0001. Control, n = 10 neurons; PSD-95 shRNA, n = 10 neurons; PSD-95-GFP-R, n = 11 neurons.

Supplementary Figure 10 D-domain and Ser332 of ephrin-B3 are required for normal localization of PSD-95.

(a) Effects of ephrin-B3 shRNA and molecular replacement of ephrin-B3 on the distribution of PSD-95-GFP in cultured cortical neurons. Representative images of PSD-95-GFP (arrows) in dendritic segments from neurons transfected with PSD-95-GFP, ephrin-B3 shRNA, and the indicated flag-tagged ephrin-B3 mutant rescue constructs. Plot profiles under the images indicate dendritic distribution of GFP fluorescence in each of the transfection conditions. The dashed line represents the intensity of GFP in dendritic shafts of control neurons (Figure 4h). Flag-ephrin-B3-L293A and phosphomimetic flag-ephrin-B3-S332D, which exhibit reduced interaction with PSD-95, do not rescue diffuse dendritic PSD-95-GFP distribution caused by the knockdown of endogenous ephrin-B3 (Figure 4 in the main text). Rescuing ephrin-B3 expression with non-phosphorylatable flag-ephrin-B-3S332A restores punctate PSD-95-GFP localization (arrows) and reduces diffuse dendritic GFP fluorescence to control levels (dotted line in profile plots). Scale bars: 3 µm.

(b) Quantification of GFP intensities per pixel in dendritic shafts. D-domain flag-ephrin-B3 (n = 10 neurons) and phosphomimetic flag-ephrin-B3_S332D (n = 10 neurons) rescue conditions exhibit higher shaft GFP fluorescence than control neurons. Non-phosphorylatable ephrin-B3-S332A mutant (n = 8 neurons) rescues shaft GFP fluorescence back to control levels. p<0.0001, F (4, 57) = 7.233, ANOVA with Tukey’s post-hoc. Control (n = 17 neurons), eB3 shRNA (n = 17 neurons) – p = 0.0074; Control, L293A (n = 10 neurons) – p = 0.0307; Control, S332D (n = 10 neurons) – p = 0.0015; Ephrin-B3 shRNA, S332A (n = 8 neurons) – p = 0.0208; L293A, S332A – p = 0.0478; S332A, S332D – p = 0.0042.

(c) Quantification of GFP intensities per pixel in puncta. ANOVA, F (4, 57) = 10.58, p < 0.0001, Tukey’s post-hoc. Control (n = 17 neurons), eB3 shRNA (n = 17 neurons) – p = 0.0003; Control, L293A (n = 10 neurons) – p = 0.0146; Control, S332D (n = 10 neurons) – p = 0.0018; Ephrin-B3 shRNA, S332A (n = 8 neurons) – p = 0.0001; L293A, S332A – p = 0.0033; S332A, S332D – p = 0.0005.

Supplementary information

Supplementary Text and Figures

Supplementary Figures 1–10 (PDF 3173 kb)

High stability of PSD-95-GFP in puncta of wild type neurons.

Representative time sequence of a FRAP experiment from a control neuron transfected with PSD-95-GFP at DIV0. Imaging was performed at DIV10 and images were acquired once every minute for 60 minutes. Bleached PSD-95-GFP puncta, shown in the white circle, exhibits poor recovery over the course of one hour, indicating low mobility of PSD-95-GFP in control neurons. For quantification see figure 5c. Scale bar: 3 μm. (MOV 264 kb)

Ephrin-B3 regulates the mobility of PSD-95 at synapses.

Representative time sequences of FRAP experiments from neurons transfected with PSD-95-GFP and either control (pSuper) or ephrin-B3 shRNA constructs at DIV0. At DIV10 neurons were loaded with FM 4-64 dye to mark pre-synaptic release sites. PSD-95-GFP puncta that co-localized with FM 4-64 dye were designated as synaptic, while PSD-95-GFP puncta that did not co-localize with FM 4-64 were designated as non-synaptic. In both transfection conditions, synaptic and non-synaptic puncta were selected for photobleaching (shown in circles) and their recovery was measured for 20 minutes, acquiring images every 20 seconds. Knockdown of endogenous ephrin-B3 resulted in increased recovery of synaptic and non-synaptic puncta compared to the control condition, indicating higher mobile PSD-95-GFP fractions in ephrin-B3 shRNA transfected neurons. Results are quantified in figure 6e, f. Scale bar: 2 μm. (MOV 2383 kb)

Stability of PSD-95 relies on the interaction with ephrin-B3.

Representative time sequences of FRAP experiments from neurons co-transfected with PSD-95-GFP and either control (pSuper) or ephrin-B3 shRNA at DIV0. Expression of ephrin-B3 after shRNA knockdown was rescued with the indicated shRNA-resistant ephrin-B3 wild type and mutant constructs that regulate interaction with PSD-95. FRAP experiments were performed at DIV10 and the recovery of bleached PSD-95-GFP (shown in circles) was followed for 20 minutes, acquiring images every 20 seconds. Knock down of ephrin-B3 resulted in higher recovery of PSD-95-GFP compared to control. Wild type flag-ephrin-B3 and non-phosphorylable flag-ephrin-B3 S332A, which show normal interaction with PSD-95, rescued PSD-95-GFP mobility to control levels. In contrast, flag-ephrin-B3 L293A and phosphomimetic flag-ephrin-B3 S332D, which exhibit reduced interaction with PSD-95, did not rescue PSD-95-GFP mobility. Thus, domains in ephrin-B3 that regulate the interaction with PSD-95 are required for the control of PSD-95 mobility. Results are quantified in figure 7b-g. Scale bar: 3 μm. (MOV 3144 kb)

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Hruska, M., Henderson, N., Xia, N. et al. Anchoring and synaptic stability of PSD-95 is driven by ephrin-B3. Nat Neurosci 18, 1594–1605 (2015). https://doi.org/10.1038/nn.4140

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1038/nn.4140

This article is cited by

Search

Advanced search

Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing